A fully polarimetric synthetic aperture radar (PolSAR) image allows the generation of a number of polarimetric descriptors. These descriptors are sensitive to changes in land use and cover. Thus, the objective of this...
详细信息
A fully polarimetric synthetic aperture radar (PolSAR) image allows the generation of a number of polarimetric descriptors. These descriptors are sensitive to changes in land use and cover. Thus, the objective of this study is twofold: first, to identify the most effective descriptors for each change type and ascertain the best complementary pairs from the selected polarimetric descriptors;and second, to develop an information fusion approach to use the unique features found in each polarimetric descriptor to obtain a better change map for urban and suburban environments. The effectiveness of each descriptor was assessed through statistical analysis of the sensitivity index in selected areas and through changedetection results obtained by using the supervised thresholding method. A good agreement was found between the statistical analysis and the performance of each descriptor. Finally, a polarimetric information fusion method based on the coupling of modified thresholding with a region-growing algorithm was implemented for the identified complementary descriptor pairs. The mapping accuracy, as measured by the Kappa coefficient, was improved by 0.09 (from 0.76 to 0.85) with a significant reduction of false and missing alarm rates compared to using single PolSAR images.
The performance of image classification usually depends on the quality of labelled datasets to be used as training samples. In the context of remote sensing, the acquisition of ground-truth data can be a difficult and...
详细信息
ISBN:
(纸本)9781479949052
The performance of image classification usually depends on the quality of labelled datasets to be used as training samples. In the context of remote sensing, the acquisition of ground-truth data can be a difficult and expensive task because it depends on the comprehensive surveys over the area of interest while the labelling task must be performed by experienced professionals. On the other hand, algorithms based on Active Learning can be helpful to overcome the lack of training samples. We present a cohesive algorithm for image classification and changedetection based on Active Learning, that tackles the lack of ground-truth data. Afterwards, we compute the Principal Component Analysis over post-classification images to detect deforestation on the eastern side of So Paulo urban area. Our approach provides a way to automatically select data samples, while it also suggests a category. The user provides the category data (labelling task) to the selected pixels which are further used as training data in the final classification step. We applied the algorithm over four 6-channels multispectral images of the Landsat 5/TM device and we classified the pixels in two categories ("forest" and "non-forest") for the years of 1986, 1996, 2003, and 2011. The changedetection, is computed through an automatic threshold applied on the post-classification images. We were able to quantify de deforestation suffered by the eastern side of Sao Paulo city along the years. Our results show that the remaining 31% of forest in 1986 reach a minimum of 25% in 2003, but afterwards it recovered to 27% of the area in 2011.
Emerging multimedia applications demands content-based video processing. Video has to be segmented into objects for content-based processing. A number of video object segmentation algorithms have been proposed such as...
详细信息
Emerging multimedia applications demands content-based video processing. Video has to be segmented into objects for content-based processing. A number of video object segmentation algorithms have been proposed such as semiautomatic and automatic. Semiautomatic methods adds burden to users and also not suitable for some applications. Automatic segmentation systems are still a challenge, although they are required by many applications. The proposed work aims at contributing to identify the gaps that are present in the current segmentation system and also to give the possible solutions to overcome those gaps so that the accurate and efficient video segmentation system can be developed.
A target can be positioned by wireless communication sensors. In practical system, the range based sensors may have biased measurements. The biases are mostly constant value, but they may jump abruptly in some special...
详细信息
ISBN:
(纸本)9781467391054
A target can be positioned by wireless communication sensors. In practical system, the range based sensors may have biased measurements. The biases are mostly constant value, but they may jump abruptly in some special scenarios. An on-line bias changedetection and estimation algorithm is presented in this paper. This algorithm can detect the jump bias based on Chi-Square Test, and then estimate the jump bias through Modified Augmented Extended Kalman filter. The feasibility and effectiveness of the proposed algorithms are illustrated in comparison with the Augmented Extended Kalman filter by simulations.
In this paper a method for moving targets detection based on Spotlight Synthetic Aperture Radar (SAR) images is proposed and tested in simulation data. To this purpose two moving target detection schemes are considere...
详细信息
In this paper a method for moving targets detection based on Spotlight Synthetic Aperture Radar (SAR) images is proposed and tested in simulation data. To this purpose two moving target detection schemes are considered: The first is based on a bank of Chirp Scaling algorithms (CSAs), and the second is based on changedetection (CD) applied to sub-images obtained by splitting the overall aperture into sub-apertures. At the next step, selecting and arranging target focusing center with a new technique based on re-centering phase computation to a reference point for each moving target. Finally Polar Format Algorithm (PFA) is applied to each re-organized raw data to obtain highly focused moving targets individually.
The use of unmanned aerial vehicles (UAVs) in civil aviation is growing up quickly, enabling new scenarios, especially in environmental monitoring and public surveillance services. So far, Earth observation has been c...
详细信息
The use of unmanned aerial vehicles (UAVs) in civil aviation is growing up quickly, enabling new scenarios, especially in environmental monitoring and public surveillance services. So far, Earth observation has been carried out only through satellite images, which are limited in resolution and suffer from important barriers such as cloud occlusion. Microdrone solutions, providing video streaming capabilities, are already available on the marketplace, but they are limited to altitudes of a few hundred feet. In contrast, UAVs equipped with high quality cameras can fly at altitudes of a few thousand feet and can fill the gap between satellite observations and ground sensors. Therefore, new needs for data processing arise, spanning from computer vision algorithms to sensor and mission management. This paper presents a solution for automatically detecting changes in images acquired at different times by patrolling UAVs flying over the same targets (but not necessarily along the same path or at the same altitude). changedetection in multi-temporal images is a prerequisite for land cover inspection, which, in turn, sets up the basis for detecting potentially dangerous or threatening situations.
Nowadays, there are applications in which the data are modelled best not as persistent tables, but rather as transient data streams. In this keynote, we discuss the limitations of current machine learning and data min...
详细信息
Nowadays, there are applications in which the data are modelled best not as persistent tables, but rather as transient data streams. In this keynote, we discuss the limitations of current machine learning and data mining algorithms. We discuss the fundamental issues in learning in dynamic environments like learning decision models that evolve over time, learning and forgetting, concept drift and changedetection. Data streams are characterized by huge amounts of data that introduce new constraints in the design of learning algorithms: limited computational resources in terms of memory, processing time and CPU power. In this talk, we present some illustrative algorithms designed to taking these constrains into account. We identify the main issues and current challenges that emerge in learning from data streams, and present open research lines for further developments.
SAR images have distinctive characteristics compared to optical images: speckle phenomenon produces strong fluctuations, and strong scatterers have radar signatures several orders of magnitude larger than others. We p...
详细信息
SAR images have distinctive characteristics compared to optical images: speckle phenomenon produces strong fluctuations, and strong scatterers have radar signatures several orders of magnitude larger than others. We propose to use an image decomposition approach to account for these peculiarities. Several methods have been proposed in the field of image processing to decompose an image into components of different nature, such as a geometrical part and a textural part. They are generally stated as an energy minimization problem where specific penalty terms are applied to each component of the sought decomposition. We decompose temporal series of SAR images into three components: speckle, strong scatterers and background. Our decomposition method is based on a discrete optimization technique by graph-cut. We apply it to changedetection tasks.
In this paper, we study the effects of shadowing-fading, electromagnetic interference and outliers on sequential algorithms in detecting spectral holes in a Cognitive Radio set up. The statistics of the primary signal...
详细信息
In this paper, we study the effects of shadowing-fading, electromagnetic interference and outliers on sequential algorithms in detecting spectral holes in a Cognitive Radio set up. The statistics of the primary signal, channel gain and the EMI are not known. Different nonparametric sequential algorithms are compared to choose appropriate algorithms to be used for energy detection and mean changedetection. Modification of a recently developed random walk test is found to work well for energy detection as well as for mean changedetection. We show via simulations and analysis that the nonparametric algorithms developed are robust to fading, EMI and outliers.
Anomaly-based Intrusion detection is a key research topic in network security due to its ability to face unknown attacks and new security threats. Moreover, new solutions should cope with scalability issues derived fr...
详细信息
Anomaly-based Intrusion detection is a key research topic in network security due to its ability to face unknown attacks and new security threats. Moreover, new solutions should cope with scalability issues derived from the growth of the Internet traffic. To this aim random aggregation through the use of sketches represents a powerful prefiltering stage that can be applied to backbone data traffic with a performance improvement wrt traditional static aggregations at subnet level. In the paper we apply the CUSUM algorithm at the bucket level to reveal the presence of anomalies in the current data and, in order to improve the detection rate, we correlate the data corresponding to traffic flows aggregation based on different fields of the network and transport level headers. As a side effect, the correlation procedure gives some hints on the typology of the intrusions since different attacks determine the variability of the statistics associated to specific header fields. The performance analysis, presented in this paper, demonstrates the effectiveness of the proposed approach, confirming the goodness of CUSUM as a change-point detection algorithm.
暂无评论